Inaccurate secondary structure predictions often indicate protein fold switching

Area of Science:

• * Biochemistry
• * Structural Biology
• * Bioinformatics

Background:

• * Most proteins follow a single structure-function model.
• * Some proteins exhibit dual structures and functions, termed
• fold-switching
• proteins.
• * Fold-switching enables protein multi-functionality and cellular process control.

Purpose of the Study:

• * To investigate the utility of secondary structure prediction discrepancies as a marker for fold-switching proteins.
• * To leverage negative information from secondary structure prediction inaccuracies.
• * To identify potential drug targets and understand uncharacterized biological processes.

Main Methods:

• * Quantified secondary structure prediction accuracies for 192 known fold-switching regions (FSRs) in the Protein Data Bank (PDB).
• * Compared prediction accuracies of FSRs with randomly selected non-fold-switching protein segments.
• * Analyzed types of secondary structure discrepancies (e.g., helix-to-strand, strand-to-coil).

Main Results:

• * Secondary structure prediction accuracies for FSRs varied significantly.
• * Inaccurate predictions strongly correlated with fold-switching proteins compared to non-fold-switching segments.
• * Inaccurate predictions were enriched in helix-to-strand and strand-to-coil discrepancies.
• * Fold-switching proteins with inaccurate predictions were often underrepresented in the PDB.

Conclusions:

• * Inconsistent secondary structure predictions serve as a preliminary marker for fold-switching proteins.
• * Unequal representation of protein conformers in the PDB may contribute to prediction inaccuracies.
• * This approach aids in identifying proteins with dual structures and functions.